The present invention relates to a method and apparatus to control heat transfer between two objects, and more specifically, a method and apparatus to control heat transfer using a system of manipulating magnetorheological fluid.
Electronic devices perform tasks, which are becoming more complicated and computationally intensive with each passing year. In response to the requirements placed on these electronic devices, semiconductor die need to perform at ever-increasing levels of performance. To provide the increased performance, successive generations of electronic devices include semiconductor die having smaller design rules which enable higher data speeds with the tradeoff of generating more heat in successively smaller spatial volumes. Further, as semiconductor die to the larger electrical device becomes more densely packed. This dense interconnection circuitry may become a physical obstacle to remove heat from the semiconductor die and contributes to the heat generated by the electrical device. Heat is often removed from the electrical device as materials making up the electrical device may be altered by temperatures above a certain threshold and these temperatures may adversely change electrical characteristics of the materials. For example, power leakage through transistors on logic circuitry may occur as the temperature is increased and data integrity issues may occur when memory cells are exposed to temperatures outside their operating range. Also, removing heat may reduce extreme temperature fluctuations in the electrical device, which can damage components through expansion and contraction when power is cycled on and off.
Conventional heat transfer approaches for semiconductor die include passive air convection, forced air conduction, and/or thermal sinks. However, these approaches are becoming less effective given the greater amounts of heat being generated in reduced spatial volumes. A known inefficiency in server and other electronic cooling is the underutilization of heat sinks based on chip usage. For example, when one processor is being used at fully capacity and another adjacent processor is not being used, the heat sink volume of the unused processor is being wasted.